Archive

An ever-growing list of blogged-upon topics…

General Practices & Random Diversions

• Improving tf-idf weighted document vector embedding
• Seeing language through character-level taggers
• Distributed Vector Representations of Folksong Motifs
• A CNN for Language-Agnostic Source Code Summarization
• NER and PoS when nothing is capitalized

Neural Machine Translation

• Massively Multilingual NMT
• Formality Style Transfer

Speech Recognition

• Making speech recognition datasets at scale with YouTube
• wav2vec: unsupervised pre-training for speech recognition

Reinforcement Learning

• Reinforcement Learned Curricula for NMT

Topic Modelling

• Re-Ranking Words to Improve Interpretability of Automatically Generated Topics

Transfer learning

• To tune or not to tune?